The science behind the PDA controversy - Part 2
This is the second part in an article series investigating the science behind Pathological Demand Avoidance (or PDA). In the first part, I talked about the controversy behind PDA. In this second part, I will go into detail about the origins of PDA and investigate whether it is even a real phenomenon. If you haven't read part 1 yet, I definitely recommend going back to it first.
Is PDA a Real Condition?
As we jump back in after discussing PDA’s controversial status, I want to start with the most basic question being asked about PDA in online spaces… Is PDA even a real condition?
Well that really depends on what you mean. PDA is clearly a real phenomena. It’s real, in that it is a profile of experience and behavior that a significant amount of people identify with, and some clinicians see as a profile that is distinct from other conditions. The experiences people who identify with PDA have are very real. But that doesn’t tell us much about what PDA is at a foundational level, or what causes it.
I haven't actually heard anyone express the opinion that PDA experiences aren’t real – but there are people who question whether it is a real syndrome, condition or disorder – and whether it is even worth talking about.
While PDA experiences are very real, PDA is not an officially recognized or diagnosable condition. It does not appear in the DSM or the ICD, so there are no official ways to be diagnosed with it. Some clinicians do offer PDA assessments with unofficial diagnoses, but there is no agreed upon criteria for PDA so these evaluations are often not consistent between clinics. Unlike conditions such as ADHD, autism, anxiety, depression or ODD, research on PDA is currently insufficient to qualify for inclusion in these diagnostic manuals. So, if by “real” you mean that it is an officially recognized condition. Then, no… it’s not.
Of course, this doesn’t mean that PDA won’t ultimately qualify for inclusion in a diagnostic manual. There is a process to being added to official diagnostic manuals like the DSM, and all the conditions currently listed in it, at some point, were still unofficial. From a scientific point of view, the current lack of rigorous research on PDA doesn’t mean that it doesn’t represent a real condition. It just means that the research isn’t sufficient right now for it to be included. With more research, this could still change.
So before we delve into the question of whether PDA should be considered it’s own official condition like autism, ADHD, anxiety, or depression, it might be helpful to ask… what makes something an official condition to begin with?
How Symptoms Become Syndromes
Psychology is a somewhat messy science, but we usually form our understanding of new conditions in a similar way. This process usually starts based on people’s complaints about unwanted symptoms or behaviors (whether those complaints come from the people with the condition or those around them). Those presenting with these clusters of symptoms are then studied by clinicians, and new disorders, syndromes or diseases are proposed. Measures and diagnostic criteria are created based on the best information scientists have at the time it’s being discussed. And these are studied to make sure that they are valid, repeatable and reliable at finding people with these clusters of symptoms.
Then those studying the cluster of symptoms submit proposals to institutions like the DSM for the proposed condition to be included. There is a rigorous gauntlet that must be passed through before a new condition is added to the DSM, including gathering specific types of evidence, showing that there is a reliable and valid way to test for the condition, and showing that it isn’t redundant to previous diagnoses – or if it is, that it's more reliable or clinically useful in some way to those diagnoses. You can read the full criteria of what you’d need to put together for the review board here.
PDA has yet to meet the criteria, in part because the needed studies haven’t been done. But research on PDA is still in an early phase of scientific development, and it is following the usual path towards recognition as a distinct condition.
PDA’s Origins
PDA was first identified and named by researcher Elizabeth Newson in the 1980’s. She noticed a profile of patients (mostly children) who presented with some, but not all, the symptoms of autism – and seemed to differ significantly in some respects from those diagnosed as autistic at that time.
The primary challenges that these children experienced was an intense resistance to the ordinary demands of everyday life, and an anxiety driven need for control. They would usually be diagnosed as having a pervasive developmental disorder –not otherwise specified (PDD-NOS), despite Newson’s belief that PDA should be its own diagnosis. Interestingly, PDD-NOS has since been folded into the autism spectrum disorders diagnosis – which may be part of why PDA is now often considered a profile of autism.
Does PDA have a Standardized Criteria?
While some research on this profile continued after its initial identification, it wasn’t until the early 2000’s that a measure was proposed for finding those with the profile, using a subset of the Diagnostic Interview for Social and Communication Disorders (DISCO). Then, in 2013, the Extreme Demand Avoidance Questionnaire (EDA-Q) was developed by Egan and O’Nions. This was designed specifically to measure PDA traits in children.
The EDA-Q assesses PDA with 26 questions for caregivers of children suspected to have PDA. It includes possible symptoms like:
Obsessively resists and avoids ordinary demands and requests
Finds everyday pressures (e.g. having to go on a school trip/visit dentist) intolerably stressful
Has difficulty complying with demands unless they are carefully presented
Has bouts of extreme emotional responses to small events (e.g. crying/giggling, becoming furious)
Social interaction has to be on his or her own terms
Attempts to negotiate better terms with adults
If the caretaker affirms enough of the right items on the list, the child meets the EDA-Q’s criteria for PDA.
The EDA-Q was initially validated as having good sensitivity and reliability in identifying those with the PDA profile. The analysis revealed that it was most likely to identify patients who were already suspected of having PDA over those in all comparison groups. This suggested it was suitable to continue using for clinical research - and most of the research done on PDA to date has been using the EDA-Q, or a modified version of it, as a measure.
In 2019, a version was created for adults, the EDA-QA, which also reported to have good enough reliability in initial studies to be considered usable for clinical research on PDA. On the face of it, this makes it look like PDA does have a validated and operationalized measure. In other words, it looks like a stable criteria for assessing if someone has PDA.
But other researchers have argued that these measures may be built on shaky foundations. The EDA-Q was compiled based on Newson’s original criteria of PDA, but we have little to no information on how Newson came up with this criteria to begin with. We know that Newson gathered data from 1975-2000 but she never published a specific methodology on how she formed the PDA criteria.
To add to this, there are circularity issues with the EDA-Q arising from the way groups in the validation studies were preselected as PDA or not, based solely on parental reports. This kind of selection bias can skew results to make the measure look more accurate than it really is. This is because the researchers selected parents who were already educated about Newson’s PDA criteria and had pre-identified their children as having PDA. These same parents were then the ones responding to the EDA-Q, on behalf of their children. This raises serious concerns that confirmation bias could be impacting the results.
Do PDA Traits Cluster Together?
Given the methodological weaknesses in the formation of the EDA-Q’s criteria, it is questionable whether it represents a cluster of symptoms that are actually statistically connected to each other in any important ways. But better statistical methods could identify whether the symptom clusters in PDA’s criteria are actually statistically associated with each other in any significant ways –or if they overlap with already established conditions in the DSM. The individual symptoms of PDA can be found in autism and other conditions, but it’s still unclear if they are clustered together as proposed by the PDA criteria.
In other words, is PDA a strong profile of symptoms clustered together, or are the boundaries of the profile too blurry to differentiate from other manifestations of autism or other conditions?
Could one person have demand avoidance, but not use social strategies? Could one have a strong need for autonomy, but not experience stress around daily chores? People that have all these symptoms certainly exist, but the symptoms themselves may occur outside the profile as regularly as within it. If this were found, it could suggest that PDA shouldn’t be considered its own diagnosis, but rather part of the range of diversity present within a larger diagnosis (like autism) or a combination of diagnoses (like autism plus ADHD and anxiety).
We don’t have research showing whether the PDA criteria represents a statistically significant cluster in the general population, but there is one study which looked for PDA traits in a group of autistic children. They found that 1 in 5 autistic children had some PDA traits, but only 1 in 25 met the full criteria for PDA. This suggests that the cluster of symptoms that make up the criteria for PDA aren’t that statistically significant as a group within autism. Instead, individual PDA traits seem to be spread out unevenly in the autistic population. If this were to be successfully replicated in other studies, it would be strong evidence that PDA is not a “profile of autism.”
Merging Pervasive Developmental Disorders
We can look at how pervasive developmental disorders (PDD’s) were condensed into autism spectrum disorders to see how this applies more broadly. It used to be that there were 5 conditions classed under PDD’s - including autism, aspergers, and PDD-NOS. But because these conditions had so many overlapping symptoms, researchers ultimately decided to collapse all five into the umbrella of autism spectrum disorders. Why? Because diagnosis wasn’t reliable and the criteria wasn’t statistically significant. Different professionals would diagnose the same people with different conditions under this umbrella because they weren’t actually distinct enough to rigorously tell the difference between them.
For instance, the boundaries between autism and aspergers were difficult to define. While there was a list of criteria for each, often those with aspergers would have symptoms from the autism list and vice versa. Statistical analysis showed that these criteria were not reflected in reality. Instead of neat groups of clustered symptoms, there was an array of different symptoms that could show up in any individual with a PDD diagnosis.
Or, some would have many symptoms of one condition, to a disabling degree, but not fit the full criteria for that diagnosis. These folks would get lumped into PDD-NOS (which was basically a catch all for those who didn’t perfectly fit under another PDD diagnosis). Those at the DSM decided it was better to have one umbrella diagnosis related to the overlapping symptoms, and then offer treatment based on the specific symptoms, associated conditions, and support needed by the individual.
It’s been suggested that PDA fits under this umbrella as well, especially now that the criteria for autism (and our understanding of the diverse ways it can present) has shifted to include a broader range of symptoms. The criteria that Newson was using to diagnose autism in the 1980’s isn’t the same criteria used today. Today, many individuals with PDA do meet the criteria for autism. But, when we consider PDA within autism, it also has these blurry boundaries like aspergers. So calling it a distinct profile of autism really doesn’t fit. Yes, some autistic people meet the full criteria, but the symptoms don’t cluster together more often than they are found individually.
Of course, it’s also notable that a recent study looking at the DSM criteria in general, found that many conditions already codified - including autism, ADHD and even depression - did not show up as natural symptom clusters when studied. Many (but not all) traits associated with autism and ADHD were correlated to a symptom cluster of neurodevelopmental differences, but weren’t distinct from each other. As science progresses, we may even see ADHD and autism collapse into one category because of these same issues - and see greater emphasis on using individual symptoms to guide treatment.
PDA was not included in that study (as it is not in the DSM), but ODD, which many point out has similarities to PDA, was actually one of the few conditions that did map onto its own statistical cluster of symptoms. The traits associated with ODD were statistically tied to each other. This is somewhat surprising, given the questionable nature of the ODD’s criteria – and widespread criticism (with strong evidence) that the diagnosis of it is racially biased. We’ll get more into that in part 4 of this series. Still, it may represent some statistically significant factor that we have yet to understand.
Until we do a similar study looking at PDA, it will be challenging to say whether it is a clinically useful category to study, or whether its symptoms are better understood as part of other conditions.
Other Methodological Worries About PDA
To add to this worry, a meta review of the PDA literature conducted in 2021 found that the current research was lacking in a few additional respects. Importantly, no studies looked at the views of individuals who had been identified as PDA. Most studies relied exclusively on parental reports for data and did not control for alternative explanations of the reported behavior, such as co-occurring anxiety or other conditions that bear similarities to PDA.
The lack of insight from individuals with PDA makes it difficult to assess the internal experience of PDA and can over-prioritize caregivers' perceptions and assumptions about their children. A lack of research on whether PDA could be explained by other conditions, also makes it particularly difficult to assess as its own condition.
When we review the criteria for PDA, many of its symptoms can be found in other conditions, which has led to the critique that it’s unnecessary to create a whole new “PDA” condition. Even worse, labeling PDA as a unique condition may actually keep people from understanding the underlying causes of their symptoms or accessing treatments already established for these other overlapping diagnoses.
So, if PDA symptoms can be explained as simply the manifestation of other conditions… what conditions fit?
The most obvious choice for an underlying condition that manifests with PDA symptoms seems to be autism, given the close ties between autism and PDA. But looking at the actual research – things are not so clear. We’ll discuss autism's connections to PDA and whether PDA is “just autism” in the next article, followed by a discussion of other connected conditions in part 4.
This article is the second in a series investigating the science behind PDA. You can find Part 3 “Is PDA Just Autism?” here.
Comments